document resume - eric · document resume. ed 097 350. ... male usaf basic airmen assigned to...
TRANSCRIPT
DOCUMENT RESUME
ED 097 350 TM 003 978
AUTHOR Valentine, Lonnie D., Jr.; Cowan, Douglas K.TITLE Comparability Study of Armed Services Vocational
Aptitude Battery Scores from Answer Sheet and AnswerCard Administration. Final Report.
INSTITUTION Air Force Human Resources Lab., Lackland AFB, Tex.Personnel Research Div.
REPORT NO AFHRL-TR-73-55PUB DATE Jan 74NOTE 16p.
EDRS PRICE MF-$0.75 HC-$1.50 PLUS POSTAGEDESCRIPTORS *Answer Sheets; *Comparative Analysis; Mil.Aary
Personnel; Occupational Tests; *Response Mode;*scores; Testing; Test Results; VocationalAptitude
IDENTIFIERS *Armed Services Vocational Aptitude Battery
ABSTRACTArmed Services Vocational Aptitude Battery (ASVAB)
Form 2 was standardized for Digitek answer sheets. In September 1973,the Digitek scoring facility at Randolph AFB will be inadequate tohandle anticipated scoring load. Consequently, ASVAB answer formshave been redesigned for processing via a Hewlett-Packard mark-sensereader from IBM card size forms which are more compressed than wasthe Digitek form. Score comparability from administrations on the twoanswer forms was investigated. It was found that the answer formchange had negligible effect on average examinee performance overallscore distributions, or correlations among the battery's tests.Significant Interaction between initial ability (as measured by theArmed Forces Qualification Test) and answer form used was found fortwo tests (Word Knowledge and Coding Speed); however the interactiondid not follow a clearcut pattern. It was recommended that existingASVAB conversion tables continue in use until completion of a fullrestandardization study. (Author)
R FORCE El
H
U
MA
AFHRL-TR-73.65
COMPARABILITY STUDY OF ARMED SERVICESVOCATIONAL APTITUDE BATTERY SCORESFROM ANSWER SHEET AND ANSWER CARD
ADMINISTRATION
By
Lonnie D. Valentine, Jr.Douglas K. Cowan
PERSONNEL RESEARCH DIVISIONLackland Air Force Base, Texas 78236
January 1974
Approved for public release; distribution unlimited.
u s DEPARTMENT or HEALTHEDUCATION & WELFARENATIONAL INSTITUTE OP
EDUCATION, , tat f 4 uft) ,' I t) f1Ar. . v F t) I LP)...
.F CA'',14P4Y r P't .4)", t ,)c) N
t; ,:r; Nr., r v.. 4 :.)( Put4N° I r N4 9',. r, F );
t-P .. r)k, p,), 4, u pc,
S LABORATORY
AIR FORCE SYSTEMS COMMANDBROOKS AIR FORCE BASE,TEXAS 78235
NOTICE
BESTCOPY
4161081E
When US Government drawings, specifications, or other data are usedfor any purpose other than a definitely related Governmentprocurement operation, the Government thereby incurs noresponsibility nor any obligation whatsoever, and the fact that theGovernment may have formulated, furnished, or in any way suppliedthe said drawings, specifications, or other data is not to be regarded byimplication or otherwise, as in any manner licensing the holder or anyother person or corporation, or conveying any rights or permission tomanufacture, use, or sell any patented invention that may in any waybe related thereto.
This final report was submitted by Personnel Research Division, AirForce Human Resources Laboratory, Lack land Air Force Base, Texas78236, under project 7719, with the Hq Air Force Human ResourcesLaboratory, Brooks Air Force Base, Texas 78235.
This report has been reviewed and cleared for open publication and/orpublic release by the appropriate Office of Information (01) inaccordance with AFR 190.17 and DoDD 5230.9. There is no objectionto unlimited distribution of this report to the public at large, or byDDC to the National Technical Information Service (NTIS).
This technical report has been reviewed and is approved.
LELAND D. BROKAW, ChiefPersonnel Research Division
HAROLD E. FISCHER, Colonel, USAFCommander
UnclassifiedSECURITY :LASSIFICATION OF THIS PAGE (When Rine Frnered)
REPORT DOCUMENTATION PAGEREAD INSTRUCTIONS
BEFORE OWL FTING FORM1 REHOR ' NUMBER
AFFIRIAR-73.55
2. GOVT ACCESSION NO. 3 RECIPIENT''., cAT aLor, NUMBER
4 TITLE !old Sublett,'COMPARABILITY STUDY OF ARMED SERVICES VOCATIONALAVM DE BATTERY SCORES FROM ANSWER SHEET ANDANSWI.R CARD ADMINISTRATION
5 TYPE OF REPORT 6 PERIOD COVERED
Final
6. PERFORMING ORG. REPORT NUMBER
7. AUTHOfq /Lonnie D.'Valentine, Jr.Douglas K. Co,,'in
6 CONTRACT OR GRANT NUMBER(5)
9 PES FnrIMIN RGANIZATION NAME AND ADDRESS
Personnel Research DivisionAir Force Human Resources LaboratoryLackland Air Force Base, Texas 78236
10. PROGRAM ELEMENT. PROJECT. TASKAREA 6 WORK UNIT NUMBERS
77191002
11. CONTROLLING OFFICE NAME AND ADDRESSFig Air Force Human Resources LaboratoryBrooks Air Force Base, Texas 78235
14. MONITORING AGENCY NAME Pt ADORESS(if different from Controlling Office)
12. REPORT DATE
January 197413 NUMBER OF PAGES
14
15 SECURITY CLASS. (of this report)
Unclassified
15a DECLASSIFICATION DOWNGRADINGSCHEDULE
16 DISTRIBUTION STATEMENT /of this Report)
Approved for public release; distribution unlimited.
17. DISTRIBUTION STATEMENT (of the almtract entered in Block 20, if different from Report)
Ie. SUPPLEMENTARY NOTES0
19. KEY 140 IDS (Continue on reverse side if nece.,saty and Identify by block number)
Armed E entices Vocational Aptitude Batteryanswer t mnatperform tnce comparabilityhigh school testing A .
20. ABSTRACT (Continue on teversu side If ne essary and identify by block number)
Armed Services Vocational Aptitude Battery (ASVAB) Form 2 was standardized for Digitek answer sheets. InSeptetnuer 1973, the Digitek scoring facility at Randolph APB will be inadequate to handle anticipated scoring load.Consequently, ASVAB answer forms have been redesigned for processing via a HewlettPackard marksatise readerfrom IBM card size forms which are more compressed than was the Digitek form. Score comparability fromadministrations on the two answer forms was investigated.
It was found that the answer form change had negligble effect on average examinee performance overall scoredistributions, or correlations among the battery's tests. Significant interaction between initial ability (as measured bythe AFQT) and answer form used was found for two tests Word Knowled e and Codin 1 Steed . hoWever the
DO "1 JAN"7 1 1473 EDITION OF 1 Nov 66 IS OBSOLETE Unclassified
SECURITY CLASSIFICATION OF THIS PAGE (When beta entered)
UnclassifiedS5GURITY CLASSIFICATION OF THIS PAGE(When Data Enter itd)
Item 20 Continued.
interaction did not follow a clearcut pattern. It was recommended that existing ASVAI3 conversion tables continuein use until completion of a full restandardization study.
UnclassifiedSECURITY CLASSIFICATION OF THIS PAGE(Wheft Data Entered)
PREFACE
This report describes effect of answer form change (from a sheet to IBM size cards)on performance on the Armed Services Vocational Aptitude Battery. Work wasaccomplished under Project 7719, Ai; Force Personnel System Development onSelection, Assignment, Evaluation, Quality Control, Retention, Promotion, andUtilization; Task 771910, Armed Forces Operational Selection Tests.
TABLE OF CONTENTS
Page
I. Introduction 5
H. Procedure 5
III. Results 5
IV. Conclusions and Recommendations 6
Refemces 14
Table
'444.04,
LIST OF TABLES
Page
1 Sample Distribution by AFQT Decile 7
2 Distribution of Educational Level 7
3 Number of Cases Reporting Completion of High School Mathematicsand Science Courses 7
4 AFQT Mental Category Distribution of Samples 7
5 Distribution of AQE Scores 8
6 Intercorrelations Among Twenty Variables 9
7 ASVAB Means and Standard Deviations for Samples 2XC and 2X 10
8 Analysis of Variance Summary 11
9 ASVAB Means and Standard Deviations with AFQT Deci les 12
10 Raw Score Distributions on Coding Speed 13
11 Raw Score Distributions for ASVAB Tests 2 Through 9 14
COMPARABILITY STUDY OF ARMED SERVICES VOCATIONAL.APTITUDE BATTERY SCORES FROM ANSWER SHEET AND
ANSWER CARL) ADMINISTRATION
I. INTRODUCTION
The Armed Services Vocational AptitudeBattery (SVAB), described by Vito la and Alley(1968) an ..1 Bayroff and Fuchs (1970), is currentlyoperation; 1 for testing of high school seniors involuntarily participating high schools throughoutthe United States. The current form of ASVABrequires the use of a DIGITEK answer sheet,scored on a DIGITEK DM100 optical scannerlocated at Randolph Air Force Base, Texas.
With only one form of the ASVAB available,test security requirements dictated that alternateforms of the test be produced. Ceinsequently,ASVAB-2 and -3 were developed and were normedusing a geographically representative sample testedat specified Armed Forces Entrance andExamining Stations (AFEES). Beginning inSeptember 1973, ASVAB-2 will be used for highschool testing, with ASVAB-3 reserved as the jointservices qualifying entrance test.
Form-2 of the ASVAB was standardized on theDIGITEK answer sheets, but with the beginning ofhigh school testing during the fall of 1973, theDIGITEK DM-100 scoring capability at RandolphAFB will he inadequate to handle the anticipatedload.
A revised answer form, printed on a three partIBM card. was designed to replace the answersheet, will, a resultant reduction in the amount ofspace allowed examinees for answer marking. Therevised form was designed for scoring via theHewlett-Packard model WAOO /D mark sensereader also at Randolph AFB.
A prat iinary study, to compare the two testresponse 1:cording methods, was conducted atLackland ifB. Texas, to determine whether therewas test ne rin slippage due to answer form change.
II. PROCIDURI.
Two matched samples. each consisting of 248male USAF basic airmen assigned to LacklandAFB for processing and training, were admin-istered Fo..in 2 of the ASVAB. One sample wasadministered Form 2XC (answer card format)while the ether was tested on ASVAB Form 2X(answer slit et format). The samples were matched
5
on Armed Forces Qualification Test (AFQT)scores contained in their records. Matching wascompleted for as many percentile scores as
available from within the total number of airmentested. Distribution of AFQT deciles for thesamples is given in Table 1.
Testing was accomplished under normal testingroom conditions. ASVAB-2XC and 2X wereadministered alternately during morning and after-noon testing sessions; each form was administeredan equal number of times during each half-day topreclude time-of-day effects. Scoring was accom-plished at Randolph AFB on the DIGITEKDM-100 for Form 2X and on the Hewlitt-Pachrdmark sense reader for Form 2XC.
Two samples were processed separately. Means,standard deviations, and Pearson Product Momentint ercorrelation matrices were computed fortwenty variables (these included years of educa-tion, AFQT, four composites from the AirmanQualifying Examination, data on completion ofspecific high school courses, and all nine tests ofthe ASVAB). In addition, each of the two sampleswas divided into eight subsamples based on AFQTdecile, and means and standard deviations of theASVAB tests were computed for each resultingsubsample. Raw score distributions on the nineASVAB tests were prepared for each of the twosamples.
For each ASVAB test, a two-way classificationanalysis of variance was computed to assess effectsof format and of the format x AFQT decile inter-action on scores. In this regard, it is noted that asignificant effect for AFQT decile would beexpected in all cases as a funrtion of correlationbetween AFQT and the test. Test norms, aspreviously established, are appropriate if in-consequential effects are found both for formatand for the interaction of AFQT decile withformat; a significant F ratio on either of theseeffects, however, would suggest normative adjust-ments as a consequence of format change.
III. RESULTS
Although the answer sheet sample (2X)contains fewer cases with education beyond highschool than does the answer card sample (2XC),the average educational levels are comparable.
BEST COPY AVAILABLE
Answer sheet sample airmen (2X) indicatedcompletion for more high school math and sciencecourses than did answer card sample cases (2XC)even though sample 2X(' had more airmencompleting 12 or more years of education (Tables2 and 31.
The two samples were matched on AFQT score;each case in sample 2X was matched with a sample2XC case with the same AFQT score. AFQT scoresappear to be fairly evenly distributed across thedeciles 20 through 90; there were no scores below20. Actual score range was 21 through 98. Enlist-ment is not authorized for mental category V( AFQT score 0 - 9) applicants, and recent policyhas also excluded those scoring below 21 onAFQT. The percentage of airmen in the samples'AFQT categories I through IV appear to be quitesimilar to the distribution for recent first termUSAF airman accessions. (Unpublished data. SeeTable 4 for AFQT category distributions.)
Further evidence of comparability of the twosamples was found in the distribution of AQEpercentile scores. While statistically significantdifferences were not found between the twogroups on any of the four AQE aptitudecomposites, there were small mean differences infavor of the answer sheet sample; the largest ofthese were on the Administrative (2.27) andElectronics (1.77) Ms. Distributions, means, andstandard deviations of the AQE composites areshown in Table 5.
Minor differences.of no significance were notedamong the intercorrelatiuns for the two samples.This serves as further evidence that format changedid not materially affect the battery (Table 6).
Table 7 shows, for both samples, means andstandard deviations for each ASVAB test. Inaddition, no significant ASVAB test meandifferences were found; the largest difference (.62)is on space perception. Thus, it can be seen thatthe format change makes no difference in averageperformance.
Outcomes of the nine two-way classificationanalysis of variance problems are summarized inTable 8. As expected, highly significant F ratioswere obtained in all cases for AFQT deck,reflecting correlation between the test and AFQT.In no instance was there a statistically significantanswer format main effect; this finding is
consistent with the quite small overall meandifferences between the two sanyies (largestdifference was .62 on space percept ion). Signifi-cant format x ANT decile effects were found fortwo of the nine tests. This interaction was
6
BEST COPY AVAILABLE
significant at the .01 level for word knowledge andat the .05 level for coding speed. Interestingly; thelargest interaction effect would have beenexpected on coding speed since it is a closelytimed test of simple clerical speed and accuracy:consequently, a change in size of the responseform had greater potential for affectingperformance of "high intelligence" and "lowintelligence" subjects differentially. An inspectionof the decileby-decile group means for the twoformat groups (Table 9) reveals that in AFQTdecile 20.29, mean coding speed performance ofthe answer sheet group is 6.16 raw score pointshigher than that of the answer card group, andthat in decile 50.59, a difference of about the.same magnitude occurs, but in favor of the answercard group. Other decile group means for codingspeed are approximately equal. In the case of theword knowledge test, format group means differ infavor of the card format group by a little morethan three raw score points in the AFQT decile50.59 and by a little over two raw score'points infavor of the answer sheet format in the AFQTdecile 60-69. Other decile group means are approximately equal.
These two significant interaction effects wouldsuggest possible differential effect of test responseformat change on test performance as a functionof level of AFQT performance. In both formatsamples, correlation between AFQT and codingspeed is quite modest (low .30s), and correlationbetween AFQT and word knowledge in bothsamples is in the low .50s. AFQT has served as thestandard against which ASVAB norms arecalibrated, but with these relatively modestcorrelations, it is possible that part of the withinAFQT decile difference washes out in conversionsutilizing the full samples (as a function of AFQTdecile distri!itional overlap). However, thiscannot be eltarly determined front these databecause of the truncation of the samples on AFQT(approximately the bottom 20 percent of thenormative reference population is nut representedhere).
Raw score distribution for both groups on allnine ASVAB tests are presented in Tables 10 and11. It is noteworthy that the ASVAB testdistributions of the two format samples differ verylittle from each other.
IV. CONCLUSIONS AND RI.COMMI.NDATIONS
The results of this.study suggest compaableASVAB score results using either the answer card
or the I )1G1.1 answer sheet. Presently estab-lished A:NABForm 2 norms may he satisfactorytier answ, card administration. Although a significant format by Al:QT decile interaction was foundon coding speed and word knowledge. overallmeans. s...indard deviations, and intercorrelationsfor the two format conditions are nearly the same.This suggests that, despite the significant inter-action or; these two tests. conversion tables maybe affected very little if at all.
Table 1. Sample Distributionby AFQT Decile
Sample*
AFQT Cecile 2XC 2X Total Percent
90.99 35 35 70 14.1180,89. 28 28 56 11.2970-79 30 30 60 12.1060.69 27 27 54 10.8950-59 33 33 66 13.314049 33 33 66 13.3130.39 37 37 74 14.91
20.29 25 25 50 10.08
Total 248 248 496 100.00
The two samples are evenly distributed across therange of AFQT scores. No score lower than 21 nor higherthan 98 was reported. The two samples arc pre.cisley matched on AFQT scores.
Table 2. Distribution of EducationalLevel
Sample*Ed ucat Ion
(years) 2 KC 2 X Total
8 3 59 4 9 13
10 23 19 4211 28 37 6512 147 145 29213 20 21 4114 15 10 2515 3 3 616 6 1 7
Tot al 248 248 496Mean 11.96 11.76 11.86S.D. 1.25 1.16 1.21
'A t t . of difference between the sample meansrevealed no agnificant difference (t = 1.83).
7
Because of AFQT truncation in the data forthis study, a final determination of whetherconversions for the ASVAB tests are affected byanswer form format change must await completionof a full standardization study. Data for such astudy have been collectedand are being analyzed.It is reconitnended that present conversions forASVAB Form 2 continue in use until completionof this standardization study.
Table 3. Number of Cases ReportingCompletion of High School Mathematics
and Science Courses
Course
Sample*
2 XC 2X
Algebra 154 162Geometry 95 106Trigonometry 29 40Chemistry 63 76Physics 34 42
'chi square tests revealed no significant differencesbetween the two groups.
Table 4. AFQT Mental CategoryDistribution of Samples
MentalCategory
Sample*
2XC 2X Total Percent*
(93.100)II
23 23 46 9.27
(65.92) 80 80 160 32.26III
(31.64) 116 116 232 46.77IV
(10.30) 29 29 58 11.69
Total 248 248 496 99.99
*Percentage of Basic Airmen in each mental categoryis similar to the percentage in each category enlisted intoUSAF during recent years (July 69 through September71).
Table 5 Distribution of AQE Scores
PercentileSample 2XC Sample 2X
Adm Elect Gen Mach Adm Elect Gen Mach
95 9 19 15 11 8 14 16 II90 2 10 6 13 7 18 8 1485 11 15 6 13 12 11 7 1980 4 20 8 5 10 22 16 1375 8 11 16 14 6 10 6 870 15 13 18 8 17 7 21 1365 17 5 15 10 12 10 16 1360 22-- 29 ...221 42 26 24 23 3255 14 31 18 35 12 35 15 2150 30 18 25 25 25 18 26 2245 21 28 33 16 20 14 35 1640 22_- 17 44 7 25 25 31 1535 16 3 11 11 21 4 13 1430 6 15 5 4 15 13 5 725 19 10 0 14 9 12 5 620 20 2 4 8 13 7 4 815 4 2 I 11 3 1 1 1010 8 3 1 , 0 6 2 0 405 0 4 0 1 0 0 0 1
01 0 2 0 0 1 1 0 1
Total 248 248 248 248 248 248 248 248Mean 49.25 56.44 56.43 55.89 51.52 58.21 57.34 56.51S.D. 21.25 21.68 18.24 21.14 21.38 21.80 19.06 22.71
8
Tab
le 6
.Ii
terc
orre
latio
ns A
mon
g T
w-l
ityV
aria
bles
*
1213
1415
1617
1819
20V
ac f
able
23
45
67
89
1011
Edu
catio
n (1
)13
36.
2235
1228
3032
0510
100
0424
12.
46.7
,/6
93:
.)31
,14
4236
3722__
4329
32-0
510
1500
0010
3543
3040
24A
FQT
Sco
re (
2143
6758
6232
5065
5068
6450
4850
3633
11'7
1052
7463
6036
5667
4668
6442
4348
2425
2629
20A
QE
Mun
n (3
)56 62
74 7733 49
53 5740 52
57 6001 05
28 3127 30
09 0908 11
26 -I,
__43 41
45 4135 40
36 452(
4-1
-
AQ
E E
lect
14)
6960
3842
6536
5454
3832
4938
3824
34-,
--,
_ ,
7172
4651
6742
6065
3845
5335
3732
463'
AQ
E. G
en (
5)62
4451
6320
4643
2617
3741
5241
4733
6549
5762
1942
4619
1933
3541
3947
26A
QE
Mec
h (6
)24
3748
5856
6157
4949
2427
1223
1630
3845
5349
6245
5552
2723
3237
21C
odin
g Sp
eed
(7)
3449
0419
2715
1,4
1839
3624
2814
3941
0629
2115
0911
2733
2637
25W
ord
Kno
wle
dge
(8)
5824
4155
3733
5129
3919
3018
5517
3648
3022
4329
2926
3417
Ari
th R
easo
ning
(9)
2956
5938
2744
4552
3142
3022
5453
2230
3643
3129
3820
Too
l Kno
wle
dge
(10)
4559
7062
4904
06-
0409
0144
5856
6353
01-
-02
0509
12Sp
ace
Perc
eptio
n (1
1)64
3735
4128
3010
2113
6433
3937
'_2
1817
1914
Mec
h C
ompr
ehen
sion
(12
)60
5957
2131
0825
1154
5760
2318
1722--
12Sh
op I
ntbr
inat
ion
(13)
6861
0205
0610
0054
5302
0204
0705
Aut
o In
form
atio
n (1
4)50
0001
1511
0058
00-
-03
0405
02E
lect
Inf
orm
atio
n (1
5)15
2415
2514
2013
1418
12A
lgeb
ra (
16)
5628
4029
Geo
met
ry (
17)
5632 46
,47 51
1, -- 3949
5637
Tri
gono
met
ry (
18)
4555
3730
Che
mis
try
(19)
49 42Ph
ysic
s (
20)
Dec
imal
poi
nts
omitt
ed. I
n ea
ch c
ell,
the
top
entr
y is
bas
ed o
n sa
mpl
e 2X
C w
hale
the
botto
m e
ntry
is b
ased
on
sam
ple
2X.
Table 7. ASVAB Means and Standard Deviationsfor Samples 2XC and 2X
Test
Sample
2XC 2XMean SD Mean SD
Coding Speed 43.60 11.42 43.87 11.95Word Knowledge 14.95 4.26 14.91 4.59Arithmetic Reasoning 15.00 5.36 14.72 5.43Tool Knowledge 15.11 5.30 15.19 5.05Space Perception 16.68 5.05 17.30 5.07Mechanical Comprehension 14.98 4.58 15.40 4.72Shop Information 15.67 4.70 15.26 4.57Automotive Information 15.99 ,1,68 16.00 4.85Electronics Information 15.06 435 15:49 4.94
10
Table 8. Analysis of Variance Summary
Source SS df MS F p SS df MS F p
Total
Coding Speed
70,401.43 495
Word Knowledge
8,712.67 495Format 18.97 1 18.97 .16 ns .20 1 .20 .017 ns
Decile 10,078,23- 1,439.75 11.80 G.001 2,755.13 7 393.59 33.05 <.001Format x
Deci le 1,721.84 7 245.98 2.02 G.05 242.25 7 34.61 2.90 <.01Error 58,582,39 480 122.05 5,715.09 480 11.91
Arithmetic Reasoning Automobile Information
Total 14,450.68 495 11,263.00 495Format 9.88 1 9.88 .60 ns .06 1 .06 .003 ns
Decile 6,336.65 7 905.24 55.06 G.001 2,540.38 7 362.91 20.20 <.001Format x
Decile 215.12 7 30.73 1.87 ns 98.14 7 14.02 .78 nsError 7,889.03 480 16.44 8,624.42 480 17.96
Electrical Information Tool Knowledge
Total 10,684.71 495 14,085.02 495Format 23.52 1 23.52 1.46 as 2.61 1 2.61 .12 ItsDocile 2,855.20 7 407.89 25.24 <.001 3,466.80 7 495.26 22.68 <.00
Format xDocile 48.24 7 6.89 .43 ns 131.97 7 18.85 .86 ns
Error 7,757.75 480 16.16 10,483.64 480 21.84
Space Perception Mechanical Comprehension
Total 12,751.97 495 10,749.56 495Format 47.82 1 47.82 3.38 ns 21.39 1 21.39 1.51 nsDocile 5,857.14 7 836.73 59.09 G.001 4,390.06 7 627.15 44.29 <.001
Format xDocile 51.21 7 7.32 .52 ns 112 02 7 16.00 1.23 ns
Error 6,795.80 480 14.16 6,226.09 480 12.97
Shop Information
Total 10,677.35 495Format 20.98 1 20.98 1.25 nsDeci le 2,494.26 7 356.32 21.15 <.001
Format xDocile 74.41 7 10.63 .63 ns
Error 8,087.70 480 16.85
11
Tab
le 9
. ASV
AB
Mea
ns a
nd S
tand
ard
Dev
iatio
ns w
ith A
FQT
Dec
des
AFQ
TE
leci
le
Cod
ing
Spee
dW
ord
Kno
wle
dge
Ari
thm
etic
Rea
soni
ngT
ool
Kno
wle
dge
Spac
ePe
rcep
tion
Mec
hani
cal
Com
preh
ensi
onSh
opin
form
atio
nA
utom
otiv
eIn
form
atio
nE
lect
roni
csIn
form
atio
n2X
C2X
2XC
2X2X
C2X
2XC
2X2X
C2X
2XC
2X2X
C2X
2XC
2X2X
C2X
90-9
9M
51.
1153
.97
17.8
318
.57
20.6
620
.43
19.4
618
.43
21.5
722
.60
19.4
919
.89
19.5
118
.31
20.0
918
.83
18.9
719
.23
S8.
8811
303.
243.
134.
193.
783.
824.
103.
113.
1133
92.
7533
72.
823.
1733
54.
074.
1980
-89
M 4
5.18
45.0
017
.29
17.7
118
3017
.96
18.2
118
4620
.96
2034
17.5
419
.11
18.6
417
.86
18.9
618
.50
17.7
919
04S
11.8
511
.98
2.56
2.39
3.74
3.99
4.27
4.63
3.04
3.21
2.94
3.17
3.07
4.64
3.58
4.83
3.14
3.50
70-7
9M
43.
5344
.37
1630
16.7
716
.93
1730
16.7
716
.53
19.1
319
5317
.20
17.2
716
.93
16.9
316
.73
1 8.
1016
.07
17.5
0S
10.0
311
.90
3.02
3.89
3.86
394
435
427
3.21
3.13
2.90
338
4.69
3.95
4.16
4.10
3.74
330
60-6
9M
44.
1947
.96
14.2
216
3314
5615
9615
.07
14.5
917
3719
.26
1533
15.8
115
5615
3714
5915
.04
14.3
014
30S
939
9.19
4.75
3.78
4.27
4.47
4.75
4.98
4.02
3.26
433
3.42
4.72
3.40
4.56
4.15
3.49
3.92
50-5
9M
46.
1840
.03
16.1
813
.09
15.2
112
.09
13.6
415
.67
1545
.16
.00
14.1
214
.40
1436
14.6
714
.94
16 2
414
.03
14.9
1S
11.0
311
.12
330
4.66
4.57
4.06
4.24
4.80
4.03
4.05
331
4.74
3.47
4.65
4.25
4.40
436
4.45
4049
M 4
2.61
40.4
514
.12
13.4
513
.42
14.2
113
.70
13.8
714
.00
14.8
814
.24
1324
14.9
113
.45
15.8
515
3013
.97
13.8
8S
12.1
612
.03
332
4.26
141
3.76
439
451
3.41
4.64
339
4.16
3.71
4.79
4.06
4.86
3.70
4.88
30-3
9M
39.
8438
.89
1230
1239
10.8
99.
9512
3111
.00
13 3
013
.24
11.9
712
.41
12.4
913
3013
.54
13.0
812
.65
12.2
4S
10.2
910
.60
4.17
437
430
4.08
5.42
5.74
4.86
4.21
3.82
3.70
450
391
4.07
431
3.18
5.41
20-2
9M
34.
2040
.36
10.2
810
.64
9.24
9.64
41.3
211
.16
1152
12.4
093
210
.84
13.0
012
0012
3212
3612
4812
.88
9.85
8.77
3.88
3.33
331
4.09
5.17
3.97
332
3.88
3.45
356
457
434
4.26
5.05
3.94
4.96
Table 10. Raw Score Distributions on Coding Speed*(Samples 2.Vand 2X)
Score 2) C 2X Score 2 XC 2X Score 2XC 2X Score 2XC 2X Score 2XC 2X
80 64 2 2 1 48 12 9 32 7 9 16
79 63 3 2 47 8 9 31 4 2 15
78 1 62 1 46 6 9 30 7 1 14
77 61 1 4 45 8 4 29 3 4 13
76 1 I 60 4 1 44 9 7 28 1 6 12
75 59 5 5 43 13 9 27 3 4 11
74 58 3 42 11 11 26 2 4 10
73 1 57 6 3 41 10 4 25 1 1 9
72 3 56 5 6 40 6 7 24 4 2 8
71 1 55 9 5 39 6 13 23 4 7
70 2 54 3 8 38 9 2 22 2 2 6
69 1 53 5 8 37 3 8 21 2 5
68 1 52 5 7 36 7 5 20 2 4
67 2 2- 5 1 9 5 35 9 8 19 2 1 3
66 2 50 6 10 34 2 7 18 I 2
65 12 49 7 8 33 4 6 17 1 2 1
Total N
'Scor range is from 0 to 100. Scoring formula: Right only
13
248 248
luhle / /. Raw Score* Distributions for ASVAB Tests 2 Through 9(Samples 2M: and 2.1)
Scores
WK AR TK Sp MC SI Al El
2XC 2X 2XC 2X 2XC 2X 2XC 2X 2XC 2X 2XC 2X 2XC 2X 2XC 2X
25 1 1 6 4 2 3 9 15 2 2 5 I 2 5 5 424 3 6 13 17 12 11 18 20 3 7 5 3 13 1 1 6 10
23 1
6 6 8 10 18 15 16 21 11 10 13 10 18 16 10 12
21 4 15 18 9 14 15 24 25 12 15 20 22-- 17 22 10 16
20 20 14 15 17 20 20 28 23 19 19 19 17 21 25 II 2219 I I 5 6 I 1 2 5 2- 1 3
18 43 30 14 12 20 18 22 19 28 32 21 25 22 14 20 23
17 19 25 24 19 21 20 17 20 21 29 31 26 26 23 27 21
16 42 30 23 18 21 25 18 22 26 30 28 39 23 21 33 2215 3 4 2 7 I 4 5 6 2 5 2 8 I 7
14 25 31 22 19 26 25 20 19 22 18 25 23 28 18 33 22
13 ,.....
, 24 17 31 IN 17 22 15 32 18 24 18 20 26 34 2912 13 14 12 14 18 18 18 10 12 9 15 9 12 12 18 13
II 2_ 7 I 4 3 4 4 3 4 2 5 I 910 16 13 18 16 18 v., 14 8 10 10 14 15 12 8 11 89 I I 7 13 II 9 10 8 9 10 13 7 9 15 18 13 10
8 6 7 II 8 14 11 5 6 11 10 5 10 9 5 8 5
7 I 5 I 3 2 I 5 3 1 3
6 5 6 4 8 2 5 5 4 4 6 5 12 4 3
5 3 3 6 4 5 1 3 3 1 4 4 3
4 1 12 3 2 2 1 1 1 3 4 2 2 1 1
3 I 1 I 1 2 1 1 1
2,- 1
,- 3 3 3 2 2 1 2 1
I I I 1 1 2 2 2
0 1
'Scoring formulae: RI3.
REFERENCES
Bayroff, A.G., & Fuchs, E.F. The Armed Services Vocational Aptitude Battery. BESRL Technical Report1161. February 1970.
Vitoin, B.M.. & Alley, W.E. Develipment and standardization of Air Force composites for the ArmedServices Vocational Aptitude Battery. AFHRLTR68.110, AD688 222. Lackland AFB, Tex.:Personnel Research Division, Air Force Human Resources Laboratory, September 1968.
14